Marvellworld Trade Ltd.

Toshiba Corporation

Marvell International Limited

Kabushiki Kaisha Toshiba

Marvell Asia Pte., Ltd.

Arvind Pruthi

Sandeep Karmarkar

Ram Kishore Johri

Shailesh P Parulekar

Mayur Shardul

Shekhar S Deshkar

Abhijeet Prakash Gole

Shailesh Shiwalkar

Premkumar Mahadev Nikam

Kanishk Rastogi

The present disclosure includes systems and techniques relating to customization of a bus adapter card. in some implementations, an apparatus includes a processor and a program memory, a bus adapter card coupled with the computing apparatus and configured to connect with a storage device, the bus adapter card computing a cache memory and a controller to cache in the cache memory data associated with the storage device, where the program memory includes a driver to communicate with the bus adapter card responsive to requests corresponding to the storage device, and the driver is configured to modify its communications with the bus adapter card responsive to information provided separate from the requests.

Customization of a bus adapter card

Systems, methods, and other embodiments associated with a lookup structure for a large block cache are described. According to one embodiment, at least two blocks of data are stored in a cache. A lookup entry is constructed that describes the at least two blocks of data. The lookup entry includes block specific information that describes individual blocks of the at least two blocks of data. The lookup entry is stored in the lookup structure.

Lookup structure for large block cache

A counting module counts a number of accesses to a first region of a storage medium during each of a first period and a second period. A count comparing module compares the number of accesses during the first period and the second period and determines whether the number of accesses during the second period is greater or less than the number of accesses during the first period by a predetermined amount. The region defining module splits the first region into two or more regions if the number of accesses during the second period is greater than the number of accesses during the first period by the predetermined amount, and merges the first region into a second region if the number of accesses during the second period is less than the number of accesses during the first period by the predetermined amount.

Hotspot detection and caching for storage devices

Systems, methods, and other embodiments associated with a distributed cache coherency protocol are described. According to one embodiment, a method includes receiving a request from a requester for access to one or more memory blocks in a block storage device that is shared by at least two physical computing machines and determining if a caching right to any of the one or more memory blocks has been granted to a different requester. If the caching right has not been granted to the different requester, access is granted to the one or more memory blocks to the requester.

Distributed cache coherency protocol

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, can perform data caching. In some implementations, a method and system include receiving information that includes a logical address, allocating a physical page in a non-volatile memory structure, mapping the logical address to a physical address of the physical page, and writing, based on the physical address, data to the non-volatile memory structure to cache information associated with the logical address. The logical address can include an identifier of a data storage device and a logical page number.

Data caching in non-volatile memory

Methods, apparatus, and systems, including computer programs encoded on a computer storage medium, manage metadata for virtual volumes. In some implementations, a method includes: loading into memory at least a portion of metadata for a virtual volume (VV) that spans data extents of different persistent storage devices, wherein the metadata comprises virtual metadata block (VMB) descriptors and virtual metadata blocks (VMBs); mapping an address of the VV to a VMB number and an index of an extent pointer within a VMB identified by the VMB number, wherein the extent pointer indicates an extent within one of the different persistent storage devices; locating a VMB descriptor in the memory based on the VMB number; and locating the identified VMB in the memory or not in the memory based on the located VMB descriptor.

Metadata management for virtual volumes

Methods, apparatus, and systems, including computer programs encoded on a computer storage medium, manage an address space. In some implementations, a method includes managing an allocation data structure for a memory, wherein the allocation data structure indicates groupings of memory space, each of the groupings having a different associated integer, and each of the groupings serving memory space in portions equal to a unit of memory space allocation times the associated integer for that grouping; receiving a request for allocation within the memory, wherein the request has an associated number of the unit of memory space allocation; and selecting one of the groupings from which to serve the request for allocation within the memory based on the associated number in comparison with values obtained using the different associated integers as an exponent.

Low overhead memory space management

A controller is communicatively coupled with a storage medium and with a cache device and configured to interface with a processor or a memory of a computer system. The controller is further configured to receive a specified quantity of address references associated with a storage system; determine a spatial distribution of references to addresses of the storage system based at least in part on the received specified quantity of the address references, where to perform a determination of the spatial distribution, the controller is further configured to omit a random quantity of address references between previous determination of the spatial distribution and said determination of the spatial distribution; combine the determined spatial distribution with the previous determination of the spatial distribution into a spatial locality metric of the storage system; and output the spatial locality metric to use in caching data from the storage medium to the cache device.

Caching based on spatial distribution of accesses to data storage devices

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for managing metadata for pools of storage disks, are described. In some implementations, a system includes a processor, a memory and a storage system. The storage system contains a storage pool including multiple storage devices. Further, the storage system is configured to store in persistent storage associated with the storage pool a bitmap that is configured to indicate whether metadata blocks corresponding to one or more virtual volumes associated with the storage pool are free or occupied. In addition, the processor is configured to perform operations including accessing at least portions of the bitmap, and managing, based on information from the accessed portions, at least some of the metadata blocks stored in the persistent storage.

Metadata management for pools of storage disks

Systems, methods, and other embodiments associated with managing memory are described. According to one embodiment, an apparatus includes a converter that dynamically converts a structure of a data representation stored in a memory, where the structure is selectively converted between a sparse format and a non-sparse format.

Conversion of in-memory data representations

Systems and techniques relating to network file systems are described. A described technique includes associating with a process that exports a network file system, where the process manages the network file system, associating with one or more peers that use the network file system, where the one or more peers cache data associated with the network file system on one or more non-volatile memory structures, receiving a file system request to read or write at least a portion of a file that is associated with the network file system, and processing the file system request by selectively communicating with the one or more peers to retrieve cached data associated with the file when the file system request indicates a read or inform the one or more peers of a write to the file when the file system request indicates a write.

Peer-to-peer based caching for network file system

The parallel RS-RAID data storage architecture can aggregate that data and checksums within each cluster into intermediate or partial sums that are transferred or distributed to other clusters. The use of intermediate data symbols, intermediate checksum symbols, cluster configuration information on the assignment of data storage devices to clusters and the operational status of data storage devices, and the like, can reduce the computational burden and latency for the error correction calculations while increasing the scalability and throughput of the parallel RS-RAID distributed data storage architecture.

Parallel Reed-Solomon RAID (RS-RAID) architecture, device, and method

The present disclosure includes systems and techniques relating to customization of a bus adapter card. In some implementations, an apparatus includes a processor and a program memory, a bus adapter card coupled with the computing apparatus and configured to connect with a storage device, the bus adapter card comprising a cache memory and a controller to cache in the cache memory data associated with the storage device, where the program memory includes a driver to communicate with the bus adapter card responsive to requests corresponding to the storage device, and the driver is configured to modify its communications with the bus adapter card responsive to information provided separate from the requests.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, manage metadata for virtual volumes. In some implementations, a method and system include defining multiple metadata blocks in a persistent storage, including information that links a virtual address space to the storage system, where the defining includes, for at least one of the multiple metadata blocks, determining multiple output addresses corresponding to the storage system, and writing the multiple output addresses and an identifier corresponding to the multiple metadata blocks in a metadata block in the persistent storage. In some implementations, a method and system include reading the multiple metadata blocks into the memory from the persistent storage, including identifying the metadata block based on the identifier; receiving an input address of the virtual address space; and obtaining a corresponding output address to the storage system using the multiple metadata blocks in the memory.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, to manage an address space for large caches are described. In some implementations, a method includes receiving data to be cached in a cache address space associated with a cache memory. The cache address space includes two or more designated portions of the cache address space, such that each of the two or more designated portions has an associated minimum amount of data allowed to be cached therein. Additionally, the respective minimum amounts of the two or more designated portions are different from each other. The method also includes selecting a cache address for caching the received data from one of the two or more designated portions of the cache address space that has an associated minimum amount of data allowed to be cached therein being less than the received data. Further, the method includes caching the received data at the selected cache address, and accessing at least a subset of the data cached at the selected cache address.

Growing community of inventors

Los Gatos, CA, United States of America

Arvind Pruthi